This study will develop an automatic system for decomposing the electromyographic signal into the constituent action potentials corresponding to the firing of individual motor units activated by motorneurons. The system will be an enhancement of a current system used over the past 20 years in many studies carried out by the Neuromuscular Research Lab at Boston University. Although the current system has been a valuable research tool, it has never been useful as a clinical tool due to limitations in processing time, accuracy and portability. Proposed enhancements will be introduced by redesigning the hardware and rewriting the decomposition software using a knowledge-based artificial intelligence language (IPUS), which has recently been developed by the team. As part of this application the enhanced system will be used in two laboratory studies and two clinical studies. The laboratory studies will investigate the modifications that occur in the firing of motor units as a function of ageing and will quantify the benefits that can be restored by exercise. The system will also be used to investigate the phenomena of motor unit substitution. The clinical studies will address the use of the device in quantifying the degree of denervation in paralyzed laryngeal muscles and in studying the effect of acute ataxia on the firing characteristics of the motorneurons in cerebellar stroke. As well as testing specific hypotheses, these studies will be used to test and improve the evolving design of the new decomposition system.